A Comparison between Practical and Theoretical Adherence to Espghan–Naspghan Guidelines about Parenteral Nutrition in the Preterm in Four European Countries

Introduction: The aim of this study was to compare the theoretical and practical adherence to the Espghan–Naspgan guidelines in the management of the parenteral nutrition (PN) in the neonatal intensive care units (NICUs) of four European countries. We aimed at assessing the clinical practices by using a clinical vignette of a very preterm infant which requires PN and to compare the results with the nutritional protocol available in the units which represents intention to treat. Previous studies suggest that clinical vignette assess more precisely practices than intention-to-treat surveys. The major interest of clinical vignettes is that they appear to be a valid and comprehensive method that directly focuses on the process of care provided in actual clinical practice. Furthermore, this is an inexpensive method for measuring the quality of care provided by a group of physicians. Vignettes are not widely used in pediatrics despite they are easily administered, less costly, and can be used in all types of clinical practices. The disadvantage of such a focus is that there may be other more important reasons for poor-quality care. Nevertheless, process measures look directly at what services are provided, whether they are provided efficiently, and whether they lead to better health. Very interestingly, the comparison of the results obtained by the clinical vignette significantly differed in many points from the results of the intent-to-treat nutritional protocol. To our knowledge, this is the first study of this type, which reports the results of a survey which aims to assess the clinical practices by using a clinical vignette of a very preterm infant which requires PN. Clearly, the physicians prescribed more appropriately in clinical practice compared with what would have been predicted based on their nutritional protocol.

Materials and Methods: A total of 199 NICUs from four European countries (55 from Germany, 45 from the UK, 49 from France, and 50 from Italy) participated in the study. This represents 74% of the neonatal units of the four countries. Physicians were presented a web-based scenario of a 28 week’s gestation infant requiring exclusive PN from birth onward (clinical case). They were then asked to prescribe PN for this patient for the first, second, and third days of life, and for the growing phase assuming that enteral feeding could not be tolerated. For each day of life, ranges with six plausible intakes and a “do not know” response were offered so that physicians can choose the most appropriate range of intake for each nutrient. In the same questionnaire, the units were asked to describe the nutritional protocols and their routine clinical practices with respect to PN (nutritional protocols) as previously described (Lapillonne et al, BMJ Open 2013). Statistical analyses were performed using the software Minitab16. We examined the null hypothesis: “no differences in proportions across theory and clinical practice.” Null hypotheses were tested using χ² test and rejected at the 5% error level.

Results and Discussion: Physicians’ behavior toward both case and theoretical questions was highly variable according to the type of macronutrient analyzed. The amino acids (AA) and lipids were initiated earlier in the clinical case than in the nutritional protocols. The initial dose of AA and glucose was not significantly different between the clinical case and the nutritional protocol, whereas the initial lipid dosage was significantly lower in the clinical case. Only the target dose of lipids was similar in the clinical case and the nutritional protocol, that of AA being significantly higher, and that of glucose being significantly lower in the clinical case. The target dose of AA was higher in the clinical case than in the nutritional protocols; the target dose of lipid was not significantly different; and the target dose of glucose was significantly lower. The comparison of obtained results with current guidelines showed that 28 to 58% of physicians did not prescribe AA, lipids, or glucose during the stable growth period at a dose which is within the recommendations demonstrating current PN practices are not adherent with guidelines. More precisely, recommended timing and dose of AA was frequently not met, the lipid administration timing varied between physicians to a scale which is even larger than that for AA. The results obtained with the clinical vignette are closer to the current recommendations for most of the parameters examined which suggest that physicians prescribed more appropriately in clinical practice compared with what would have been predicted based on their nutritional protocol. This finding was unexpected, but we may postulate that since nutrition guidelines are rapidly changing, clinical practices may also change more quickly over time than the in-house protocols which may require additional work for the rewriting ([Table 1]).

Keywords: comparison, theoretical vs. practical, Espghan–Naspgan guidelines, parenteral nutrition

Table 1 Comparison between macronutrient intakes assessed by using a clinical vignette (i.e., clinical case) and those obtained by using an intention to treat questionnaire (i.e., nutritional protocol)

Nutrient	In analysis	Clinical case	Nutritional protocol	p-Value
Amino acids	Initiation time
	D0	74%	63%	<0.05
	> D0	26%	37%
	Initial dose (g/kg/d)
	≤ 1	58%	62%	NS
	> 1	42%	38%
	Target dose (g/kg/d)
	≤ 3	38%	52.5%	<0.05
	> 3	60%	47.5%
Glucose	Initial dose (g/kg/d)
	≤ 6	36%	45%	NS
	> 6	64%	55%
	Target dose (g/kg/d)
	≤ 12	45%	0	<0.01
	> 12	53%	94%
Lipid	Initiation time
	D0	36%	20%	<0.01
	D1	44%	48%
	D2	17%	22%
	D3 or >	3%	10%
	Initial dose(g/kg/d)
	≤ 1	90%	98%	<0.01
	> 1	10%	2%
	Target dose (g/kg/d)
	< 3	20%	21%	NS
	≥ 3	80%	79%